Refactoring 20,000+ URLs and the SEO impact

[jhdavids8]

I run a site that is largely powered by user reviews. We have almost 20,000 reviews, and each review has its own unique URL (/items/item-reviewed/reviews/1), as each one is quite lengthy and detailed (much longer than the normal Yelp review). Of course, the item being reviewed has its own URL (/items/item-reviewed), and we would very much prefer users are driven to that page rather than a review page in search results.

I've been looking into ways to improve our SEO, and I'm wondering if the current structure is hurting our SEO to the item page, and if so, what is the best way to 'solve' the issue without causing future SEO issues. Basically, are the 20,000 (and growing) review pages reducing the SEO impact of the actual item pages? I'd like to get the content in the reviews indexed, but not at the expense of negative SEO impact on the items being reviewed.

I have several follow-up questions if the answer to my question is indeed 'Yes, it is negatively impacting the SEO of your item page', so I'll await a response. Thanks!

[Wi-Phye]

I think the answer to this one is pretty easy. Just look at every other e-com site with reviews. They are all on the same page as the product. For usability sake the review are usually hidden in some way on the same page.

Tab, accordian, read more links etc.. When it comes to this type of question always follow the SEO masters at amazon.

I also suggest you mark up the reviews in a micro format. Not for SEO but for click thru rates.

[jhdavids8]

Thanks for the response. I was fairly sure that was the answer but wanted to be sure before I littered the post with conditional follow-ups. To be clear, the reviews are really, really long, and easily make up their own page. There are usually about 20-40 questions (with answers ranging from text to a star rating) per review, so following Amazon is easier said then done, but I want to make sure we're taking the best possible route. The item page also gives review previews for each review, with a "Read more" link for each review that takes them to the review page.

That said, here are my followups:

• In order to get the entire review indexed, as you said, I could hide the review on the page. But isn't that an SEO no-no, as Google could interpret such a large amount of hidden divs and content incorrectly? To get past that before, I've usually shown the review on initial page load and hide what I want with Javascript afterwards. Would that be a better solution?

• So say I do indeed get rid of the reviews as their own page and instead open up the review in a lightbox modal when "Read more" is clicked (my current plan). Now, I have 20,000 indexed URLs I need to do something with to avoid 20,000 404s. The way I see it, I can do one of the 2 in order to maintain a URL that gets the user to the review they want on the item page:

1. Setup the review modals to match a url param (/items/item-reviewed?reviews=1)
2. Setup the review modals to match a url anchor (/items/item-reviewed#reviews=1)

I'll of course want to 301 redirect the previous review URLs. If I chose option #1, wouldn't I still have the same issue, as Google would still index the URL with a query parameter separate from the item URL, right? However, if I went option #2, could I even 301 to the new anchor URL? I know the anchor is client-side only, but after some research, it sounds like everything but IE would support a redirect to a URL with an anchor. In this case, does Google just treat the 301 as a redirection to the item page, practically ignoring the anchor? Are there any negative SEO impacts option #2 presents (apart from IE stripping the anchor on the redirection)? Would (assume an item has 30 reviews) 30 permanent redirects to a single URL be perfectly fine via SEO standards?

Thanks so much for your time!

[Wi-Phye]

I didn't realize the reviews were that long. That does kind of present a problem and yes you don't want to hide them in a non-display element. I have used css z-index and slide the review into the viewport instead of using javascript to switch the display attribute. But I have only done this on a few small sites. Nothing like yours. The reviews were of limited length so that worked for me. Probably won't work in your case.

Hum, interesting problem. You said you already have a preview and read more link on the product page but the review page out ranks the product page? Or is this just theoretical?

Sounds like you need to do some A /  B  testing to find out why and then promote the preforming content to the main page.

Stupid question but, if your review page is already highly ranked why not add the product to each of those? I have to assume there is an add to cart function on each page.

This seem like less of an SEO question then one about conversion rates. Which is OK. That's the whole point.

[jhdavids8]

The issue is largely theoretical. The product pages seem to usually outrank the review pages, but I'm just wondering that with so many links on the item pages directing to even more content, are both competing with each other in SEO results and could the item pages' SEO be improved (even if it isn't that bad at the moment) by simply having one page for search engines to focus on?

As for adding the product to each of the reviews, we do indeed do this in a limited manner. I provide breadcrumbs to show the user where they are from a site structure layout as well as a few details on the item itself (as well as our own version of 'add to cart'), but that's it.

Alongside the potential SEO impact, I gotta think that providing some way to view the review on-page (lightbox modal) would stil benefit from a user experience. Taking them away from the item page to a review page and hoping they hit the back button is probably something we should address. Now, as you said, how I handle that is less of an SEO issue, but the potential elimination of all those review URLs is, so I'm wondering also how to handle the 404s and 301s if I go this route. Like you said, interesting issue

Again, thanks for all the help!